A new search engine called Cuil is getting a lot of attention today, primarily due to one of the founders Anna Patterson and her previous employer.  On Cuil’s management page we see that Anna is quite the hard core search geek:

Anna’s focus is on scaling architecture, tackling one of the major problems in search—the exponential growth of the Internet. Anna was the architect of Google’s large search index, TeraGoogle, that launched in early 2006. While at Google, Anna was the technical lead of one of the two Web ranking groups at Google, in charge of GoogleBase, and the manager for the core piece of Google’s ad-matching technology. She joined Google in 2004 after designing, writing and selling Recall—the largest search engine in existence at the time at 12 billion pages. Anna has a PhD in Computer Science from the University of Illinois at Urbana-Champaign, and was a Research Scientist at Stanford University.

As I mentioned, the new search engine is getting a ton of press so if you want a full scale review feel free to read these articles:

• Fortune: Cuil not a Google killer - yet
• Time: Why Cuil Is No Threat to Google
• PC World: Cuil Stumbles out of the Gate
• ZDNet: Ex-Googlers launch Cuil, but maybe too early?

Instead, I’m going to take a look at a few queries and point out some areas where I think Cuil could use some work.

Clustering

Cuil isn’t doing a great job of clustering the results and as a result you’ll see that many of the top search results are littered with listings from the same site. For example, a search for “hybrid cars” shows 5 listings above the fold. Of these, 3 of the listings are from the same website.

Number of Results - Exclusive vs. Inclusive

I like the related query tabs and the collapsable category links. So from the above query I clicked on “hybrid cars for sale“, since it is a query I monitor often. The first thing I noticed is that it appends the new phrase instead of starting a new search. This could’ve produced worse results since it refines the set further but the results are the same with and without the original query.

I understand this is their claim to fame at the moment, but sometimes in life less is more. For example, the “hybrid cars for sale” query produces 1,765,337,909 results. The same query on Google produces 3,220,000 results. We know that the vast majority of users don’t go past the first page of results. Most people don’t even make it past the first ranking. So why should we care about the other millions of results? The quality at the top is more important. The search engine should be more exclusive than inclusive in order to retain the high quality results it needs to contend with Google. I left Yahoo & MSN out because they aren’t really contenders in the search space.

Lack of Spam Filters

Do a query for some of porn, pills, casino queries.  Yikes.

Lack of Emphasis on Links / Authority

As shown in the rankings above and most of the other queries I tried there are a lot of low quality sites at the top of the results.  These could probably be eliminated by implementing more weight on inbound links, the quality of them, quantity, link text etc.

It’s almost time for the NFL season, so I did a search for “Patriots“.  Patriots.com, ESPN.com, SI.com, Foxsports.com, Yahoo.com, Boston.com, NFL.com, and a handful of other highly relevant, authoritative sites are not in the top 10.   This was a pretty bad set of results.

Incorrect Thumbnails w/Listings

I’m from Massachusetts and a huge Red Sox fan.  So I did a search for “red sox“.  This is the thumbnail that was displayed with the listing.  Oh no, no, no no.

That’s a lot of negative, but there are a lot of things I like about the search engine.  I’ll post an update shortly.